search

zylo117 / Yet-Another-EfficientDet-Pytorch

The pytorch re-implement of the official efficientdet with SOTA performance in real time and pretrained weights.
GNU Lesser General Public License v3.0
5.2k stars 1.27k forks source link

regression_loss #728

Open jzpcn opened 2 years ago

jzpcn commented 2 years ago

regression_loss = torch.where( torch.le(regression_diff, 1.0 / 9.0), 0.5 9.0 torch.pow(regression_diff, 2), regression_diff - 0.5 / 9.0 )

why you use this method to compute reg_loss rather than the normal smooth L1 loss